2 * chimeraslayercommand.cpp
5 * Created by westcott on 3/31/10.
6 * Copyright 2010 Schloss Lab. All rights reserved.
10 #include "chimeraslayercommand.h"
11 #include "deconvolutecommand.h"
12 #include "referencedb.h"
13 #include "sequenceparser.h"
15 //**********************************************************************************************************************
16 vector<string> ChimeraSlayerCommand::setParameters(){
18 CommandParameter ptemplate("reference", "InputTypes", "", "", "none", "none", "none",false,true); parameters.push_back(ptemplate);
19 CommandParameter pfasta("fasta", "InputTypes", "", "", "none", "none", "none",false,true); parameters.push_back(pfasta);
20 CommandParameter pname("name", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(pname);
21 CommandParameter pgroup("group", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(pgroup);
22 CommandParameter pwindow("window", "Number", "", "50", "", "", "",false,false); parameters.push_back(pwindow);
23 CommandParameter pksize("ksize", "Number", "", "7", "", "", "",false,false); parameters.push_back(pksize);
24 CommandParameter pmatch("match", "Number", "", "5.0", "", "", "",false,false); parameters.push_back(pmatch);
25 CommandParameter pmismatch("mismatch", "Number", "", "-4.0", "", "", "",false,false); parameters.push_back(pmismatch);
26 CommandParameter pminsim("minsim", "Number", "", "90", "", "", "",false,false); parameters.push_back(pminsim);
27 CommandParameter pmincov("mincov", "Number", "", "70", "", "", "",false,false); parameters.push_back(pmincov);
28 CommandParameter pminsnp("minsnp", "Number", "", "10", "", "", "",false,false); parameters.push_back(pminsnp);
29 CommandParameter pminbs("minbs", "Number", "", "90", "", "", "",false,false); parameters.push_back(pminbs);
30 CommandParameter psearch("search", "Multiple", "kmer-blast", "blast", "", "", "",false,false); parameters.push_back(psearch);
31 CommandParameter pprocessors("processors", "Number", "", "1", "", "", "",false,false); parameters.push_back(pprocessors);
32 CommandParameter prealign("realign", "Boolean", "", "T", "", "", "",false,false); parameters.push_back(prealign);
33 CommandParameter ptrim("trim", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(ptrim);
34 CommandParameter psplit("split", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(psplit);
35 CommandParameter pnumwanted("numwanted", "Number", "", "15", "", "", "",false,false); parameters.push_back(pnumwanted);
36 CommandParameter piters("iters", "Number", "", "1000", "", "", "",false,false); parameters.push_back(piters);
37 CommandParameter pdivergence("divergence", "Number", "", "1.007", "", "", "",false,false); parameters.push_back(pdivergence);
38 CommandParameter pparents("parents", "Number", "", "3", "", "", "",false,false); parameters.push_back(pparents);
39 CommandParameter pincrement("increment", "Number", "", "5", "", "", "",false,false); parameters.push_back(pincrement);
40 CommandParameter pblastlocation("blastlocation", "String", "", "", "", "", "",false,false); parameters.push_back(pblastlocation);
41 CommandParameter pinputdir("inputdir", "String", "", "", "", "", "",false,false); parameters.push_back(pinputdir);
42 CommandParameter poutputdir("outputdir", "String", "", "", "", "", "",false,false); parameters.push_back(poutputdir);
43 CommandParameter psave("save", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(psave);
45 vector<string> myArray;
46 for (int i = 0; i < parameters.size(); i++) { myArray.push_back(parameters[i].name); }
50 m->errorOut(e, "ChimeraSlayerCommand", "setParameters");
54 //**********************************************************************************************************************
55 string ChimeraSlayerCommand::getHelpString(){
57 string helpString = "";
58 helpString += "The chimera.slayer command reads a fastafile and referencefile and outputs potentially chimeric sequences.\n";
59 helpString += "This command was modeled after the chimeraSlayer written by the Broad Institute.\n";
60 helpString += "The chimera.slayer command parameters are fasta, name, template, processors, trim, ksize, window, match, mismatch, divergence. minsim, mincov, minbs, minsnp, parents, search, iters, increment, numwanted, blastlocation and realign.\n";
61 helpString += "The fasta parameter allows you to enter the fasta file containing your potentially chimeric sequences, and is required, unless you have a valid current fasta file. \n";
62 helpString += "The name parameter allows you to provide a name file, if you are using reference=self. \n";
63 helpString += "The group parameter allows you to provide a group file. The group file can be used with a namesfile and reference=self. When checking sequences, only sequences from the same group as the query sequence will be used as the reference. \n";
64 helpString += "You may enter multiple fasta files by separating their names with dashes. ie. fasta=abrecovery.fasta-amazon.fasta \n";
65 helpString += "The reference parameter allows you to enter a reference file containing known non-chimeric sequences, and is required. You may also set template=self, in this case the abundant sequences will be used as potential parents. \n";
66 helpString += "The processors parameter allows you to specify how many processors you would like to use. The default is 1. \n";
68 helpString += "When using MPI, the processors parameter is set to the number of MPI processes running. \n";
70 helpString += "The trim parameter allows you to output a new fasta file containing your sequences with the chimeric ones trimmed to include only their longest piece, default=F. \n";
71 helpString += "The split parameter allows you to check both pieces of non-chimeric sequence for chimeras, thus looking for trimeras and quadmeras. default=F. \n";
72 helpString += "The window parameter allows you to specify the window size for searching for chimeras, default=50. \n";
73 helpString += "The increment parameter allows you to specify how far you move each window while finding chimeric sequences, default=5.\n";
74 helpString += "The numwanted parameter allows you to specify how many sequences you would each query sequence compared with, default=15.\n";
75 helpString += "The ksize parameter allows you to input kmersize, default is 7, used if search is kmer. \n";
76 helpString += "The match parameter allows you to reward matched bases in blast search, default is 5. \n";
77 helpString += "The parents parameter allows you to select the number of potential parents to investigate from the numwanted best matches after rating them, default is 3. \n";
78 helpString += "The mismatch parameter allows you to penalize mismatched bases in blast search, default is -4. \n";
79 helpString += "The divergence parameter allows you to set a cutoff for chimera determination, default is 1.007. \n";
80 helpString += "The iters parameter allows you to specify the number of bootstrap iters to do with the chimeraslayer method, default=1000.\n";
81 helpString += "The minsim parameter allows you to specify a minimum similarity with the parent fragments, default=90. \n";
82 helpString += "The mincov parameter allows you to specify minimum coverage by closest matches found in template. Default is 70, meaning 70%. \n";
83 helpString += "The minbs parameter allows you to specify minimum bootstrap support for calling a sequence chimeric. Default is 90, meaning 90%. \n";
84 helpString += "The minsnp parameter allows you to specify percent of SNPs to sample on each side of breakpoint for computing bootstrap support (default: 10) \n";
85 helpString += "The search parameter allows you to specify search method for finding the closest parent. Choices are blast, and kmer, default blast. \n";
86 helpString += "The realign parameter allows you to realign the query to the potential parents. Choices are true or false, default true. \n";
87 helpString += "The blastlocation parameter allows you to specify the location of your blast executable. By default mothur will look in ./blast/bin relative to mothur's executable. \n";
88 helpString += "If the save parameter is set to true the reference sequences will be saved in memory, to clear them later you can use the clear.memory command. Default=f.";
89 helpString += "The chimera.slayer command should be in the following format: \n";
90 helpString += "chimera.slayer(fasta=yourFastaFile, reference=yourTemplate, search=yourSearch) \n";
91 helpString += "Example: chimera.slayer(fasta=AD.align, reference=core_set_aligned.imputed.fasta, search=kmer) \n";
92 helpString += "Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFastaFile).\n";
96 m->errorOut(e, "ChimeraSlayerCommand", "getHelpString");
100 //**********************************************************************************************************************
101 ChimeraSlayerCommand::ChimeraSlayerCommand(){
103 abort = true; calledHelp = true;
105 vector<string> tempOutNames;
106 outputTypes["chimera"] = tempOutNames;
107 outputTypes["accnos"] = tempOutNames;
108 outputTypes["fasta"] = tempOutNames;
110 catch(exception& e) {
111 m->errorOut(e, "ChimeraSlayerCommand", "ChimeraSlayerCommand");
115 //***************************************************************************************************************
116 ChimeraSlayerCommand::ChimeraSlayerCommand(string option) {
118 abort = false; calledHelp = false;
119 ReferenceDB* rdb = ReferenceDB::getInstance();
121 //allow user to run help
122 if(option == "help") { help(); abort = true; calledHelp = true; }
123 else if(option == "citation") { citation(); abort = true; calledHelp = true;}
126 vector<string> myArray = setParameters();
128 OptionParser parser(option);
129 map<string,string> parameters = parser.getParameters();
131 ValidParameters validParameter("chimera.slayer");
132 map<string,string>::iterator it;
134 //check to make sure all parameters are valid for command
135 for (it = parameters.begin(); it != parameters.end(); it++) {
136 if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
139 vector<string> tempOutNames;
140 outputTypes["chimera"] = tempOutNames;
141 outputTypes["accnos"] = tempOutNames;
142 outputTypes["fasta"] = tempOutNames;
144 //if the user changes the input directory command factory will send this info to us in the output parameter
145 string inputDir = validParameter.validFile(parameters, "inputdir", false);
146 if (inputDir == "not found"){ inputDir = ""; }
148 //check for required parameters
149 fastafile = validParameter.validFile(parameters, "fasta", false);
150 if (fastafile == "not found") {
151 //if there is a current fasta file, use it
152 string filename = m->getFastaFile();
153 if (filename != "") { fastaFileNames.push_back(filename); m->mothurOut("Using " + filename + " as input file for the fasta parameter."); m->mothurOutEndLine(); }
154 else { m->mothurOut("You have no current fastafile and the fasta parameter is required."); m->mothurOutEndLine(); abort = true; }
156 m->splitAtDash(fastafile, fastaFileNames);
158 //go through files and make sure they are good, if not, then disregard them
159 for (int i = 0; i < fastaFileNames.size(); i++) {
162 if (fastaFileNames[i] == "current") {
163 fastaFileNames[i] = m->getFastaFile();
164 if (fastaFileNames[i] != "") { m->mothurOut("Using " + fastaFileNames[i] + " as input file for the fasta parameter where you had given current."); m->mothurOutEndLine(); }
166 m->mothurOut("You have no current fastafile, ignoring current."); m->mothurOutEndLine(); ignore=true;
167 //erase from file list
168 fastaFileNames.erase(fastaFileNames.begin()+i);
175 if (inputDir != "") {
176 string path = m->hasPath(fastaFileNames[i]);
177 //if the user has not given a path then, add inputdir. else leave path alone.
178 if (path == "") { fastaFileNames[i] = inputDir + fastaFileNames[i]; }
184 ableToOpen = m->openInputFile(fastaFileNames[i], in, "noerror");
186 //if you can't open it, try default location
187 if (ableToOpen == 1) {
188 if (m->getDefaultPath() != "") { //default path is set
189 string tryPath = m->getDefaultPath() + m->getSimpleName(fastaFileNames[i]);
190 m->mothurOut("Unable to open " + fastaFileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine();
192 ableToOpen = m->openInputFile(tryPath, in2, "noerror");
194 fastaFileNames[i] = tryPath;
198 if (ableToOpen == 1) {
199 if (m->getOutputDir() != "") { //default path is set
200 string tryPath = m->getOutputDir() + m->getSimpleName(fastaFileNames[i]);
201 m->mothurOut("Unable to open " + fastaFileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine();
203 ableToOpen = m->openInputFile(tryPath, in2, "noerror");
205 fastaFileNames[i] = tryPath;
211 if (ableToOpen == 1) {
212 m->mothurOut("Unable to open " + fastaFileNames[i] + ". It will be disregarded."); m->mothurOutEndLine();
213 //erase from file list
214 fastaFileNames.erase(fastaFileNames.begin()+i);
217 m->setFastaFile(fastaFileNames[i]);
222 //make sure there is at least one valid file left
223 if (fastaFileNames.size() == 0) { m->mothurOut("[ERROR]: no valid files."); m->mothurOutEndLine(); abort = true; }
227 //check for required parameters
229 namefile = validParameter.validFile(parameters, "name", false);
230 if (namefile == "not found") { namefile = ""; hasName = false; }
232 m->splitAtDash(namefile, nameFileNames);
234 //go through files and make sure they are good, if not, then disregard them
235 for (int i = 0; i < nameFileNames.size(); i++) {
238 if (nameFileNames[i] == "current") {
239 nameFileNames[i] = m->getNameFile();
240 if (nameFileNames[i] != "") { m->mothurOut("Using " + nameFileNames[i] + " as input file for the name parameter where you had given current."); m->mothurOutEndLine(); }
242 m->mothurOut("You have no current namefile, ignoring current."); m->mothurOutEndLine(); ignore=true;
243 //erase from file list
244 nameFileNames.erase(nameFileNames.begin()+i);
251 if (inputDir != "") {
252 string path = m->hasPath(nameFileNames[i]);
253 //if the user has not given a path then, add inputdir. else leave path alone.
254 if (path == "") { nameFileNames[i] = inputDir + nameFileNames[i]; }
260 ableToOpen = m->openInputFile(nameFileNames[i], in, "noerror");
262 //if you can't open it, try default location
263 if (ableToOpen == 1) {
264 if (m->getDefaultPath() != "") { //default path is set
265 string tryPath = m->getDefaultPath() + m->getSimpleName(nameFileNames[i]);
266 m->mothurOut("Unable to open " + nameFileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine();
268 ableToOpen = m->openInputFile(tryPath, in2, "noerror");
270 nameFileNames[i] = tryPath;
274 if (ableToOpen == 1) {
275 if (m->getOutputDir() != "") { //default path is set
276 string tryPath = m->getOutputDir() + m->getSimpleName(nameFileNames[i]);
277 m->mothurOut("Unable to open " + nameFileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine();
279 ableToOpen = m->openInputFile(tryPath, in2, "noerror");
281 nameFileNames[i] = tryPath;
287 if (ableToOpen == 1) {
288 m->mothurOut("Unable to open " + nameFileNames[i] + ". It will be disregarded."); m->mothurOutEndLine();
289 //erase from file list
290 nameFileNames.erase(nameFileNames.begin()+i);
293 m->setNameFile(nameFileNames[i]);
298 //make sure there is at least one valid file left
299 if (nameFileNames.size() == 0) { m->mothurOut("[ERROR]: no valid name files."); m->mothurOutEndLine(); abort = true; }
302 if (hasName && (nameFileNames.size() != fastaFileNames.size())) { m->mothurOut("[ERROR]: The number of namefiles does not match the number of fastafiles, please correct."); m->mothurOutEndLine(); abort=true; }
304 bool hasGroup = true;
305 groupfile = validParameter.validFile(parameters, "group", false);
306 if (groupfile == "not found") { groupfile = ""; hasGroup = false; }
308 m->splitAtDash(groupfile, groupFileNames);
310 //go through files and make sure they are good, if not, then disregard them
311 for (int i = 0; i < groupFileNames.size(); i++) {
314 if (groupFileNames[i] == "current") {
315 groupFileNames[i] = m->getGroupFile();
316 if (groupFileNames[i] != "") { m->mothurOut("Using " + groupFileNames[i] + " as input file for the group parameter where you had given current."); m->mothurOutEndLine(); }
318 m->mothurOut("You have no current namefile, ignoring current."); m->mothurOutEndLine(); ignore=true;
319 //erase from file list
320 groupFileNames.erase(groupFileNames.begin()+i);
327 if (inputDir != "") {
328 string path = m->hasPath(groupFileNames[i]);
329 //if the user has not given a path then, add inputdir. else leave path alone.
330 if (path == "") { groupFileNames[i] = inputDir + groupFileNames[i]; }
336 ableToOpen = m->openInputFile(groupFileNames[i], in, "noerror");
338 //if you can't open it, try default location
339 if (ableToOpen == 1) {
340 if (m->getDefaultPath() != "") { //default path is set
341 string tryPath = m->getDefaultPath() + m->getSimpleName(groupFileNames[i]);
342 m->mothurOut("Unable to open " + groupFileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine();
344 ableToOpen = m->openInputFile(tryPath, in2, "noerror");
346 groupFileNames[i] = tryPath;
350 if (ableToOpen == 1) {
351 if (m->getOutputDir() != "") { //default path is set
352 string tryPath = m->getOutputDir() + m->getSimpleName(groupFileNames[i]);
353 m->mothurOut("Unable to open " + groupFileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine();
355 ableToOpen = m->openInputFile(tryPath, in2, "noerror");
357 groupFileNames[i] = tryPath;
363 if (ableToOpen == 1) {
364 m->mothurOut("Unable to open " + groupFileNames[i] + ". It will be disregarded."); m->mothurOutEndLine();
365 //erase from file list
366 groupFileNames.erase(groupFileNames.begin()+i);
369 m->setGroupFile(groupFileNames[i]);
374 //make sure there is at least one valid file left
375 if (groupFileNames.size() == 0) { m->mothurOut("[ERROR]: no valid group files."); m->mothurOutEndLine(); abort = true; }
378 if (hasGroup && (groupFileNames.size() != fastaFileNames.size())) { m->mothurOut("[ERROR]: The number of groupfiles does not match the number of fastafiles, please correct."); m->mothurOutEndLine(); abort=true; }
381 //if the user changes the output directory command factory will send this info to us in the output parameter
382 outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ outputDir = ""; }
384 string temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = m->getProcessors(); }
385 m->setProcessors(temp);
386 convert(temp, processors);
388 temp = validParameter.validFile(parameters, "save", false); if (temp == "not found"){ temp = "f"; }
389 save = m->isTrue(temp);
391 if (save) { //clear out old references
396 it = parameters.find("reference");
397 //user has given a template file
398 if(it != parameters.end()){
399 if (it->second == "self") {
400 templatefile = "self";
402 m->mothurOut("[WARNING]: You can't save reference=self, ignoring save.");
403 m->mothurOutEndLine();
408 path = m->hasPath(it->second);
409 //if the user has not given a path then, add inputdir. else leave path alone.
410 if (path == "") { parameters["reference"] = inputDir + it->second; }
412 templatefile = validParameter.validFile(parameters, "reference", true);
413 if (templatefile == "not open") { abort = true; }
414 else if (templatefile == "not found") { //check for saved reference sequences
415 if (rdb->referenceSeqs.size() != 0) {
416 templatefile = "saved";
418 m->mothurOut("[ERROR]: You don't have any saved reference sequences and the reference parameter is a required.");
419 m->mothurOutEndLine();
422 }else { if (save) { rdb->setSavedReference(templatefile); } }
424 }else if (hasName) { templatefile = "self";
426 m->mothurOut("[WARNING]: You can't save reference=self, ignoring save.");
427 m->mothurOutEndLine();
432 if (rdb->referenceSeqs.size() != 0) {
433 templatefile = "saved";
435 m->mothurOut("[ERROR]: You don't have any saved reference sequences and the reference parameter is a required.");
436 m->mothurOutEndLine();
437 templatefile = ""; abort = true;
443 temp = validParameter.validFile(parameters, "ksize", false); if (temp == "not found") { temp = "7"; }
444 convert(temp, ksize);
446 temp = validParameter.validFile(parameters, "window", false); if (temp == "not found") { temp = "50"; }
447 convert(temp, window);
449 temp = validParameter.validFile(parameters, "match", false); if (temp == "not found") { temp = "5"; }
450 convert(temp, match);
452 temp = validParameter.validFile(parameters, "mismatch", false); if (temp == "not found") { temp = "-4"; }
453 convert(temp, mismatch);
455 temp = validParameter.validFile(parameters, "divergence", false); if (temp == "not found") { temp = "1.007"; }
458 temp = validParameter.validFile(parameters, "minsim", false); if (temp == "not found") { temp = "90"; }
459 convert(temp, minSimilarity);
461 temp = validParameter.validFile(parameters, "mincov", false); if (temp == "not found") { temp = "70"; }
462 convert(temp, minCoverage);
464 temp = validParameter.validFile(parameters, "minbs", false); if (temp == "not found") { temp = "90"; }
465 convert(temp, minBS);
467 temp = validParameter.validFile(parameters, "minsnp", false); if (temp == "not found") { temp = "10"; }
468 convert(temp, minSNP);
470 temp = validParameter.validFile(parameters, "parents", false); if (temp == "not found") { temp = "3"; }
471 convert(temp, parents);
473 temp = validParameter.validFile(parameters, "realign", false); if (temp == "not found") { temp = "t"; }
474 realign = m->isTrue(temp);
476 temp = validParameter.validFile(parameters, "trim", false); if (temp == "not found") { temp = "f"; }
477 trim = m->isTrue(temp);
479 temp = validParameter.validFile(parameters, "split", false); if (temp == "not found") { temp = "f"; }
480 trimera = m->isTrue(temp);
482 search = validParameter.validFile(parameters, "search", false); if (search == "not found") { search = "blast"; }
484 temp = validParameter.validFile(parameters, "iters", false); if (temp == "not found") { temp = "1000"; }
485 convert(temp, iters);
487 temp = validParameter.validFile(parameters, "increment", false); if (temp == "not found") { temp = "5"; }
488 convert(temp, increment);
490 temp = validParameter.validFile(parameters, "numwanted", false); if (temp == "not found") { temp = "15"; }
491 convert(temp, numwanted);
493 blastlocation = validParameter.validFile(parameters, "blastlocation", false);
494 if (blastlocation == "not found") { blastlocation = ""; }
496 //add / to name if needed
497 string lastChar = blastlocation.substr(blastlocation.length()-1);
498 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
499 if (lastChar != "/") { blastlocation += "/"; }
501 if (lastChar != "\\") { blastlocation += "\\"; }
503 blastlocation = m->getFullPathName(blastlocation);
504 string formatdbCommand = "";
505 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
506 formatdbCommand = blastlocation + "formatdb";
508 formatdbCommand = blastlocation + "formatdb.exe";
511 //test to make sure formatdb exists
513 formatdbCommand = m->getFullPathName(formatdbCommand);
514 int ableToOpen = m->openInputFile(formatdbCommand, in, "no error"); in.close();
515 if(ableToOpen == 1) { m->mothurOut("[ERROR]: " + formatdbCommand + " file does not exist. mothur requires formatdb.exe to run chimera.slayer."); m->mothurOutEndLine(); abort = true; }
517 string blastCommand = "";
518 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
519 blastCommand = blastlocation + "megablast";
521 blastCommand = blastlocation + "megablast.exe";
523 //test to make sure formatdb exists
525 blastCommand = m->getFullPathName(blastCommand);
526 ableToOpen = m->openInputFile(blastCommand, in2, "no error"); in2.close();
527 if(ableToOpen == 1) { m->mothurOut("[ERROR]: " + blastCommand + " file does not exist. mothur requires blastall.exe to run chimera.slayer."); m->mothurOutEndLine(); abort = true; }
530 if ((search != "blast") && (search != "kmer")) { m->mothurOut(search + " is not a valid search."); m->mothurOutEndLine(); abort = true; }
532 if (hasName && (templatefile != "self")) { m->mothurOut("You have provided a namefile and the reference parameter is not set to self. I am not sure what reference you are trying to use, aborting."); m->mothurOutEndLine(); abort=true; }
533 if (hasGroup && (templatefile != "self")) { m->mothurOut("You have provided a group file and the reference parameter is not set to self. I am not sure what reference you are trying to use, aborting."); m->mothurOutEndLine(); abort=true; }
535 //until we resolve the issue 10-18-11
536 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
542 catch(exception& e) {
543 m->errorOut(e, "ChimeraSlayerCommand", "ChimeraSlayerCommand");
547 //***************************************************************************************************************
549 int ChimeraSlayerCommand::execute(){
551 if (abort == true) { if (calledHelp) { return 0; } return 2; }
553 for (int s = 0; s < fastaFileNames.size(); s++) {
555 m->mothurOut("Checking sequences from " + fastaFileNames[s] + " ..." ); m->mothurOutEndLine();
557 int start = time(NULL);
559 //you provided a groupfile
560 string groupFile = "";
561 if (groupFileNames.size() != 0) { groupFile = groupFileNames[s]; }
563 //maps a filename to priority map.
564 //if no groupfile this is fastafileNames[s] -> prioirity
565 //if groupfile then this is each groups seqs -> priority
566 map<string, map<string, int> > fileToPriority;
567 map<string, map<string, int> >::iterator itFile;
568 map<string, string> fileGroup;
569 fileToPriority[fastaFileNames[s]] = priority; //default
570 fileGroup[fastaFileNames[s]] = "noGroup";
571 SequenceParser* parser = NULL;
573 int totalChimeras = 0;
575 if ((templatefile == "self") && (groupFile == "")) {
577 fileToPriority.clear();
578 if (processors != 1) { m->mothurOut("When using template=self, mothur can only use 1 processor, continuing."); m->mothurOutEndLine(); processors = 1; }
579 string nameFile = "";
580 if (nameFileNames.size() != 0) { //you provided a namefile and we don't need to create one
581 nameFile = nameFileNames[s];
582 }else { nameFile = getNamesFile(fastaFileNames[s]); }
584 //sort fastafile by abundance, returns new sorted fastafile name
585 m->mothurOut("Sorting fastafile according to abundance..."); cout.flush();
586 priority = sortFastaFile(fastaFileNames[s], nameFile);
587 m->mothurOut("Done."); m->mothurOutEndLine();
589 fileToPriority.clear();
591 fileToPriority[fastaFileNames[s]] = priority;
592 fileGroup[fastaFileNames[s]] = "noGroup";
593 if (m->control_pressed) { for (int j = 0; j < outputNames.size(); j++) { m->mothurRemove(outputNames[j]); } return 0; }
594 }else if ((templatefile == "self") && (groupFile != "")) {
596 fileToPriority.clear();
597 if (processors != 1) { m->mothurOut("When using template=self, mothur can only use 1 processor, continuing."); m->mothurOutEndLine(); processors = 1; }
598 string nameFile = "";
599 if (nameFileNames.size() != 0) { //you provided a namefile and we don't need to create one
600 nameFile = nameFileNames[s];
601 }else { nameFile = getNamesFile(fastaFileNames[s]); }
603 //Parse sequences by group
604 parser = new SequenceParser(groupFile, fastaFileNames[s], nameFile);
605 vector<string> groups = parser->getNamesOfGroups();
607 for (int i = 0; i < groups.size(); i++) {
608 vector<Sequence> thisGroupsSeqs = parser->getSeqs(groups[i]);
609 map<string, string> thisGroupsMap = parser->getNameMap(groups[i]);
610 string newFastaFile = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + groups[i] + "-sortedTemp.fasta";
611 priority = sortFastaFile(thisGroupsSeqs, thisGroupsMap, newFastaFile);
612 fileToPriority[newFastaFile] = priority;
613 fileGroup[newFastaFile] = groups[i];
617 if (outputDir == "") { outputDir = m->hasPath(fastaFileNames[s]); }//if user entered a file with a path then preserve it
618 string outputFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "slayer.chimera";
619 string accnosFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "slayer.accnos";
620 string trimFastaFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "slayer.fasta";
623 ofstream out, out1, out2;
624 m->openOutputFile(outputFileName, out); out.close();
625 m->openOutputFile(accnosFileName, out1); out1.close();
626 if (trim) { m->openOutputFile(trimFastaFileName, out2); out2.close(); }
627 outputNames.push_back(outputFileName); outputTypes["chimera"].push_back(outputFileName);
628 outputNames.push_back(accnosFileName); outputTypes["accnos"].push_back(accnosFileName);
629 if (trim) { outputNames.push_back(trimFastaFileName); outputTypes["fasta"].push_back(trimFastaFileName); }
632 for (itFile = fileToPriority.begin(); itFile != fileToPriority.end(); itFile++) {
634 string thisFastaName = itFile->first;
635 map<string, int> thisPriority = itFile->second;
637 //this is true when you have parsed by groups
638 if (fileToPriority.size() > 1) { m->mothurOutEndLine(); m->mothurOut("Checking sequences from group: " + fileGroup[thisFastaName] + "."); m->mothurOutEndLine(); }
640 string thisoutputFileName = outputDir + m->getRootName(m->getSimpleName(thisFastaName)) + fileGroup[thisFastaName] + "slayer.chimera";
641 string thisaccnosFileName = outputDir + m->getRootName(m->getSimpleName(thisFastaName)) + fileGroup[thisFastaName] + "slayer.accnos";
642 string thistrimFastaFileName = outputDir + m->getRootName(m->getSimpleName(thisFastaName)) + fileGroup[thisFastaName] + "slayer.fasta";
644 //create chimera here if you are mac or linux because fork will copy for you. Create in create processes instead if you are windows.
645 if (processors == 1) { templateSeqsLength = setupChimera(thisFastaName, thisPriority); }
647 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || USE_MPI
648 templateSeqsLength = setupChimera(thisFastaName, thisPriority);
652 if (m->control_pressed) { if (parser != NULL) { delete parser; } for (int j = 0; j < outputNames.size(); j++) { m->mothurRemove(outputNames[j]); } return 0; }
655 MPIExecute(thisFastaName, thisoutputFileName, thisaccnosFileName, thistrimFastaFileName);
656 if (m->control_pressed) { outputTypes.clear(); for (int j = 0; j < outputNames.size(); j++) { m->mothurRemove(outputNames[j]); } return 0; }
659 vector<unsigned long long> positions;
660 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
661 positions = m->divideFile(thisFastaName, processors);
662 for (int i = 0; i < (positions.size()-1); i++) { lines.push_back(new linePair(positions[i], positions[(i+1)])); }
664 if (processors == 1) { lines.push_back(new linePair(0, 1000)); }
666 positions = m->setFilePosFasta(thisFastaName, numSeqs);
668 //figure out how many sequences you have to process
669 int numSeqsPerProcessor = numSeqs / processors;
670 for (int i = 0; i < processors; i++) {
671 int startIndex = i * numSeqsPerProcessor;
672 if(i == (processors - 1)){ numSeqsPerProcessor = numSeqs - i * numSeqsPerProcessor; }
673 lines.push_back(new linePair(positions[startIndex], numSeqsPerProcessor));
679 numSeqs = driver(lines[0], thisoutputFileName, thisFastaName, thisaccnosFileName, thistrimFastaFileName);
681 int numNoParents = chimera->getNumNoParents();
682 if (numNoParents == numSeqs) { m->mothurOut("[WARNING]: megablast returned 0 potential parents for all your sequences. This could be due to formatdb.exe not being setup properly, please check formatdb.log for errors."); m->mothurOutEndLine(); }
684 }else{ numSeqs = createProcesses(thisoutputFileName, thisFastaName, thisaccnosFileName, thistrimFastaFileName); }
686 if (m->control_pressed) { if (parser != NULL) { delete parser; } outputTypes.clear(); if (trim) { m->mothurRemove(trimFastaFileName); } m->mothurRemove(outputFileName); m->mothurRemove(accnosFileName); for (int j = 0; j < outputNames.size(); j++) { m->mothurRemove(outputNames[j]); } for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); delete chimera; return 0; }
690 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || USE_MPI
694 for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear();
697 m->appendFiles(thisoutputFileName, outputFileName); m->mothurRemove(thisoutputFileName);
698 totalChimeras = m->appendFiles(thisaccnosFileName, accnosFileName); m->mothurRemove(thisaccnosFileName);
699 if (trim) { m->appendFiles(thistrimFastaFileName, trimFastaFileName); m->mothurRemove(thistrimFastaFileName); }
701 totalSeqs += numSeqs;
704 if (fileToPriority.size() > 1) { totalChimeras = deconvoluteResults(parser, outputFileName, accnosFileName, trimFastaFileName); }
706 if (parser != NULL) { delete parser; }
708 m->mothurOutEndLine(); m->mothurOut(toString(totalChimeras) + " chimera found."); m->mothurOutEndLine(); m->mothurOut("It took " + toString(time(NULL) - start) + " secs to check " + toString(totalSeqs) + " sequences."); m->mothurOutEndLine();
711 //set accnos file as new current accnosfile
713 itTypes = outputTypes.find("accnos");
714 if (itTypes != outputTypes.end()) {
715 if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setAccnosFile(current); }
719 itTypes = outputTypes.find("fasta");
720 if (itTypes != outputTypes.end()) {
721 if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setFastaFile(current); }
725 m->mothurOutEndLine();
726 m->mothurOut("Output File Names: "); m->mothurOutEndLine();
727 for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
728 m->mothurOutEndLine();
733 catch(exception& e) {
734 m->errorOut(e, "ChimeraSlayerCommand", "execute");
738 //**********************************************************************************************************************
739 int ChimeraSlayerCommand::MPIExecute(string inputFile, string outputFileName, string accnosFileName, string trimFastaFileName){
743 int pid, numSeqsPerProcessor;
745 vector<unsigned long long> MPIPos;
748 MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are
749 MPI_Comm_size(MPI_COMM_WORLD, &processors);
753 MPI_File outMPIAccnos;
754 MPI_File outMPIFasta;
756 int outMode=MPI_MODE_CREATE|MPI_MODE_WRONLY;
757 int inMode=MPI_MODE_RDONLY;
759 char outFilename[1024];
760 strcpy(outFilename, outputFileName.c_str());
762 char outAccnosFilename[1024];
763 strcpy(outAccnosFilename, accnosFileName.c_str());
765 char outFastaFilename[1024];
766 strcpy(outFastaFilename, trimFastaFileName.c_str());
768 char inFileName[1024];
769 strcpy(inFileName, inputFile.c_str());
771 MPI_File_open(MPI_COMM_WORLD, inFileName, inMode, MPI_INFO_NULL, &inMPI); //comm, filename, mode, info, filepointer
772 MPI_File_open(MPI_COMM_WORLD, outFilename, outMode, MPI_INFO_NULL, &outMPI);
773 MPI_File_open(MPI_COMM_WORLD, outAccnosFilename, outMode, MPI_INFO_NULL, &outMPIAccnos);
774 if (trim) { MPI_File_open(MPI_COMM_WORLD, outFastaFilename, outMode, MPI_INFO_NULL, &outMPIFasta); }
776 if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); if (trim) { MPI_File_close(&outMPIFasta); } MPI_File_close(&outMPIAccnos); delete chimera; return 0; }
778 if (pid == 0) { //you are the root process
779 m->mothurOutEndLine();
780 m->mothurOut("Only reporting sequence supported by " + toString(minBS) + "% of bootstrapped results.");
781 m->mothurOutEndLine();
783 string outTemp = "Name\tLeftParent\tRightParent\tDivQLAQRB\tPerIDQLAQRB\tBootStrapA\tDivQLBQRA\tPerIDQLBQRA\tBootStrapB\tFlag\tLeftWindow\tRightWindow\n";
786 int length = outTemp.length();
787 char* buf2 = new char[length];
788 memcpy(buf2, outTemp.c_str(), length);
790 MPI_File_write_shared(outMPI, buf2, length, MPI_CHAR, &status);
793 MPIPos = m->setFilePosFasta(inputFile, numSeqs); //fills MPIPos, returns numSeqs
795 if (templatefile != "self") { //if template=self we can only use 1 processor
796 //send file positions to all processes
797 for(int i = 1; i < processors; i++) {
798 MPI_Send(&numSeqs, 1, MPI_INT, i, tag, MPI_COMM_WORLD);
799 MPI_Send(&MPIPos[0], (numSeqs+1), MPI_LONG, i, tag, MPI_COMM_WORLD);
802 //figure out how many sequences you have to align
803 numSeqsPerProcessor = numSeqs / processors;
804 int startIndex = pid * numSeqsPerProcessor;
805 if(pid == (processors - 1)){ numSeqsPerProcessor = numSeqs - pid * numSeqsPerProcessor; }
807 if (templatefile == "self") { //if template=self we can only use 1 processor
809 numSeqsPerProcessor = numSeqs;
813 driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, outMPIAccnos, outMPIFasta, MPIPos);
815 int numNoParents = chimera->getNumNoParents();
817 for(int i = 1; i < processors; i++) {
818 MPI_Recv(&temp, 1, MPI_INT, 1, tag, MPI_COMM_WORLD, &status);
819 numNoParents += temp;
823 if (numSeqs == numNoParents) { m->mothurOut("[WARNING]: megablast returned 0 potential parents for all your sequences. This could be due to formatdb.exe not being setup properly, please check formatdb.log for errors."); m->mothurOutEndLine(); }
825 if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); if (trim) { MPI_File_close(&outMPIFasta); } MPI_File_close(&outMPIAccnos); delete chimera; return 0; }
827 }else{ //you are a child process
828 if (templatefile != "self") { //if template=self we can only use 1 processor
829 MPI_Recv(&numSeqs, 1, MPI_INT, 0, tag, MPI_COMM_WORLD, &status);
830 MPIPos.resize(numSeqs+1);
831 MPI_Recv(&MPIPos[0], (numSeqs+1), MPI_LONG, 0, tag, MPI_COMM_WORLD, &status);
833 //figure out how many sequences you have to align
834 numSeqsPerProcessor = numSeqs / processors;
835 int startIndex = pid * numSeqsPerProcessor;
836 if(pid == (processors - 1)){ numSeqsPerProcessor = numSeqs - pid * numSeqsPerProcessor; }
839 driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, outMPIAccnos, outMPIFasta, MPIPos);
841 int numNoParents = chimera->getNumNoParents();
842 MPI_Send(&numNoParents, 1, MPI_INT, 0, tag, MPI_COMM_WORLD);
844 if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); if (trim) { MPI_File_close(&outMPIFasta); } MPI_File_close(&outMPIAccnos); delete chimera; return 0; }
850 MPI_File_close(&inMPI);
851 MPI_File_close(&outMPI);
852 MPI_File_close(&outMPIAccnos);
853 if (trim) { MPI_File_close(&outMPIFasta); }
854 MPI_Barrier(MPI_COMM_WORLD); //make everyone wait - just in case
860 catch(exception& e) {
861 m->errorOut(e, "ChimeraSlayerCommand", "MPIExecute");
865 //**********************************************************************************************************************
866 int ChimeraSlayerCommand::deconvoluteResults(SequenceParser* parser, string outputFileName, string accnosFileName, string trimFileName){
868 map<string, string> uniqueNames = parser->getAllSeqsMap();
869 map<string, string>::iterator itUnique;
874 m->openInputFile(accnosFileName, in2, "no error");
877 m->openOutputFile(accnosFileName+".temp", out2);
879 string name; name = "";
880 set<string> chimerasInFile;
881 set<string>::iterator itChimeras;
884 if (m->control_pressed) { in2.close(); out2.close(); m->mothurRemove(outputFileName); m->mothurRemove((accnosFileName+".temp")); return 0; }
886 in2 >> name; m->gobble(in2);
889 itUnique = uniqueNames.find(name);
891 if (itUnique == uniqueNames.end()) { m->mothurOut("[ERROR]: trouble parsing accnos results. Cannot find "+ name + "."); m->mothurOutEndLine(); m->control_pressed = true; }
893 itChimeras = chimerasInFile.find((itUnique->second));
895 if (itChimeras == chimerasInFile.end()) {
896 out2 << itUnique->second << endl;
897 chimerasInFile.insert((itUnique->second));
905 m->mothurRemove(accnosFileName);
906 rename((accnosFileName+".temp").c_str(), accnosFileName.c_str());
911 m->openInputFile(outputFileName, in);
914 m->openOutputFile(outputFileName+".temp", out); out.setf(ios::fixed, ios::floatfield); out.setf(ios::showpoint);
916 string rest, parent1, parent2, line;
917 set<string> namesInFile; //this is so if a sequence is found to be chimera in several samples we dont write it to the results file more than once
918 set<string>::iterator itNames;
920 //assumptions - in file each read will always look like...
923 F11Fcsw_63104 F11Fcsw_33372 F11Fcsw_37007 0.89441 80.4469 0.2 1.03727 93.2961 52.2 no 0-241 243-369
928 line = m->getline(in); m->gobble(in);
932 //for the chimera file, we want to make sure if any group finds a sequence to be chimeric then all groups do,
933 //so if this is a report that did not find it to be chimeric, but it appears in the accnos file,
934 //then ignore this report and continue until we find the report that found it to be chimeric
938 if (m->control_pressed) { in.close(); out.close(); m->mothurRemove((outputFileName+".temp")); return 0; }
940 in >> name; m->gobble(in);
941 in >> parent1; m->gobble(in);
943 if (name == "Name") { //name = "Name" because we append the header line each time we add results from the groups
944 line = m->getline(in); m->gobble(in);
946 if (parent1 == "no") {
948 itUnique = uniqueNames.find(name);
950 if (itUnique == uniqueNames.end()) { m->mothurOut("[ERROR]: trouble parsing chimera results. Cannot find "+ name + "."); m->mothurOutEndLine(); m->control_pressed = true; }
952 //is this sequence really not chimeric??
953 itChimeras = chimerasInFile.find(itUnique->second);
955 if (itChimeras == chimerasInFile.end()) {
956 itNames = namesInFile.find((itUnique->second));
958 if (itNames == namesInFile.end()) {cout << itUnique->second << endl; out << itUnique->second << '\t' << "no" << endl; namesInFile.insert(itUnique->second); }
961 }else { //read the rest of the line
962 double DivQLAQRB,PerIDQLAQRB,BootStrapA,DivQLBQRA,PerIDQLBQRA,BootStrapB;
963 string flag, range1, range2;
965 in >> parent2 >> DivQLAQRB >> PerIDQLAQRB >> BootStrapA >> DivQLBQRA >> PerIDQLBQRA >> BootStrapB >> flag >> range1 >> range2; m->gobble(in);
968 itUnique = uniqueNames.find(name);
970 if (itUnique == uniqueNames.end()) { m->mothurOut("[ERROR]: trouble parsing chimera results. Cannot find "+ name + "."); m->mothurOutEndLine(); m->control_pressed = true; }
972 name = itUnique->second;
973 //is this name already in the file
974 itNames = namesInFile.find((name));
976 if (itNames == namesInFile.end()) { //no not in file
977 if (flag == "no") { //are you really a no??
978 //is this sequence really not chimeric??
979 itChimeras = chimerasInFile.find(name);
981 //then you really are a no so print, otherwise skip
982 if (itChimeras == chimerasInFile.end()) { print = true; }
983 }else{ print = true; }
990 namesInFile.insert(name);
992 //output parent1's name
993 itUnique = uniqueNames.find(parent1);
994 if (itUnique == uniqueNames.end()) { m->mothurOut("[ERROR]: trouble parsing chimera results. Cannot find parentA "+ parent1 + "."); m->mothurOutEndLine(); m->control_pressed = true; }
995 else { out << itUnique->second << '\t'; }
997 //output parent2's name
998 itUnique = uniqueNames.find(parent2);
999 if (itUnique == uniqueNames.end()) { m->mothurOut("[ERROR]: trouble parsing chimera results. Cannot find parentA "+ parent2 + "."); m->mothurOutEndLine(); m->control_pressed = true; }
1000 else { out << itUnique->second << '\t'; }
1002 out << DivQLAQRB << '\t' << PerIDQLAQRB << '\t' << BootStrapA << '\t' << DivQLBQRA << '\t' << PerIDQLBQRA << '\t' << BootStrapB << '\t' << flag << '\t' << range1 << '\t' << range2 << endl;
1010 m->mothurRemove(outputFileName);
1011 rename((outputFileName+".temp").c_str(), outputFileName.c_str());
1016 m->openInputFile(trimFileName, in3);
1019 m->openOutputFile(trimFileName+".temp", out3);
1021 namesInFile.clear();
1023 while (!in3.eof()) {
1024 if (m->control_pressed) { in3.close(); out3.close(); m->mothurRemove(outputFileName); m->mothurRemove(accnosFileName); m->mothurRemove((trimFileName+".temp")); return 0; }
1026 Sequence seq(in3); m->gobble(in3);
1028 if (seq.getName() != "") {
1030 itUnique = uniqueNames.find(seq.getName());
1032 if (itUnique == uniqueNames.end()) { m->mothurOut("[ERROR]: trouble parsing accnos results. Cannot find "+ seq.getName() + "."); m->mothurOutEndLine(); m->control_pressed = true; }
1034 itNames = namesInFile.find((itUnique->second));
1036 if (itNames == namesInFile.end()) {
1037 seq.printSequence(out3);
1045 m->mothurRemove(trimFileName);
1046 rename((trimFileName+".temp").c_str(), trimFileName.c_str());
1051 catch(exception& e) {
1052 m->errorOut(e, "ChimeraSlayerCommand", "deconvoluteResults");
1056 //**********************************************************************************************************************
1057 int ChimeraSlayerCommand::setupChimera(string inputFile, map<string, int>& priority){
1059 if (templatefile != "self") { //you want to run slayer with a reference template
1060 chimera = new ChimeraSlayer(inputFile, templatefile, trim, search, ksize, match, mismatch, window, divR, minSimilarity, minCoverage, minBS, minSNP, parents, iters, increment, numwanted, realign, blastlocation, rand());
1062 chimera = new ChimeraSlayer(inputFile, templatefile, trim, priority, search, ksize, match, mismatch, window, divR, minSimilarity, minCoverage, minBS, minSNP, parents, iters, increment, numwanted, realign, blastlocation, rand());
1065 if (m->control_pressed) { delete chimera; return 0; }
1067 if (chimera->getUnaligned()) { delete chimera; m->mothurOut("Your template sequences are different lengths, please correct."); m->mothurOutEndLine(); m->control_pressed = true; return 0; }
1069 return (chimera->getLength());
1071 catch(exception& e) {
1072 m->errorOut(e, "ChimeraSlayerCommand", "setupChimera");
1076 //**********************************************************************************************************************
1078 string ChimeraSlayerCommand::getNamesFile(string& inputFile){
1080 string nameFile = "";
1082 m->mothurOutEndLine(); m->mothurOut("No namesfile given, running unique.seqs command to generate one."); m->mothurOutEndLine(); m->mothurOutEndLine();
1084 //use unique.seqs to create new name and fastafile
1085 string inputString = "fasta=" + inputFile;
1086 m->mothurOut("/******************************************/"); m->mothurOutEndLine();
1087 m->mothurOut("Running command: unique.seqs(" + inputString + ")"); m->mothurOutEndLine();
1089 Command* uniqueCommand = new DeconvoluteCommand(inputString);
1090 uniqueCommand->execute();
1092 map<string, vector<string> > filenames = uniqueCommand->getOutputFiles();
1094 delete uniqueCommand;
1096 m->mothurOut("/******************************************/"); m->mothurOutEndLine();
1098 nameFile = filenames["name"][0];
1099 inputFile = filenames["fasta"][0];
1103 catch(exception& e) {
1104 m->errorOut(e, "ChimeraSlayerCommand", "getNamesFile");
1108 //**********************************************************************************************************************
1110 int ChimeraSlayerCommand::driver(linePair* filePos, string outputFName, string filename, string accnos, string fasta){
1113 m->openOutputFile(outputFName, out);
1116 m->openOutputFile(accnos, out2);
1119 if (trim) { m->openOutputFile(fasta, out3); }
1122 m->openInputFile(filename, inFASTA);
1124 inFASTA.seekg(filePos->start);
1126 if (filePos->start == 0) { chimera->printHeader(out); }
1133 if (m->control_pressed) { out.close(); out2.close(); if (trim) { out3.close(); } inFASTA.close(); return 1; }
1135 Sequence* candidateSeq = new Sequence(inFASTA); m->gobble(inFASTA);
1136 string candidateAligned = candidateSeq->getAligned();
1138 if (candidateSeq->getName() != "") { //incase there is a commented sequence at the end of a file
1139 if (candidateSeq->getAligned().length() != templateSeqsLength) {
1140 m->mothurOut(candidateSeq->getName() + " is not the same length as the template sequences. Skipping."); m->mothurOutEndLine();
1143 chimera->getChimeras(candidateSeq);
1145 if (m->control_pressed) { delete candidateSeq; return 1; }
1147 //if you are not chimeric, then check each half
1148 data_results wholeResults = chimera->getResults();
1150 //determine if we need to split
1151 bool isChimeric = false;
1153 if (wholeResults.flag == "yes") {
1154 string chimeraFlag = "no";
1155 if( (wholeResults.results[0].bsa >= minBS && wholeResults.results[0].divr_qla_qrb >= divR)
1157 (wholeResults.results[0].bsb >= minBS && wholeResults.results[0].divr_qlb_qra >= divR) ) { chimeraFlag = "yes"; }
1160 if (chimeraFlag == "yes") {
1161 if ((wholeResults.results[0].bsa >= minBS) || (wholeResults.results[0].bsb >= minBS)) { isChimeric = true; }
1165 if ((!isChimeric) && trimera) {
1167 //split sequence in half by bases
1168 string leftQuery, rightQuery;
1169 Sequence tempSeq(candidateSeq->getName(), candidateAligned);
1170 divideInHalf(tempSeq, leftQuery, rightQuery);
1172 //run chimeraSlayer on each piece
1173 Sequence* left = new Sequence(candidateSeq->getName(), leftQuery);
1174 Sequence* right = new Sequence(candidateSeq->getName(), rightQuery);
1177 chimera->getChimeras(left);
1178 data_results leftResults = chimera->getResults();
1180 chimera->getChimeras(right);
1181 data_results rightResults = chimera->getResults();
1183 //if either piece is chimeric then report
1184 Sequence trimmed = chimera->print(out, out2, leftResults, rightResults);
1185 if (trim) { trimmed.printSequence(out3); }
1187 delete left; delete right;
1189 }else { //already chimeric
1191 Sequence trimmed = chimera->print(out, out2);
1192 if (trim) { trimmed.printSequence(out3); }
1200 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
1201 unsigned long long pos = inFASTA.tellg();
1202 if ((pos == -1) || (pos >= filePos->end)) { break; }
1204 if (inFASTA.eof()) { break; }
1207 delete candidateSeq;
1209 if((count) % 100 == 0){ m->mothurOut("Processing sequence: " + toString(count)); m->mothurOutEndLine(); }
1212 if((count) % 100 != 0){ m->mothurOut("Processing sequence: " + toString(count)); m->mothurOutEndLine(); }
1216 if (trim) { out3.close(); }
1221 catch(exception& e) {
1222 m->errorOut(e, "ChimeraSlayerCommand", "driver");
1226 //**********************************************************************************************************************
1228 int ChimeraSlayerCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File& outMPI, MPI_File& outAccMPI, MPI_File& outFastaMPI, vector<unsigned long long>& MPIPos){
1232 MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are
1234 for(int i=0;i<num;i++){
1236 if (m->control_pressed) { return 1; }
1238 //read next sequence
1239 int length = MPIPos[start+i+1] - MPIPos[start+i];
1241 char* buf4 = new char[length];
1242 MPI_File_read_at(inMPI, MPIPos[start+i], buf4, length, MPI_CHAR, &status);
1244 string tempBuf = buf4;
1245 if (tempBuf.length() > length) { tempBuf = tempBuf.substr(0, length); }
1246 istringstream iss (tempBuf,istringstream::in);
1250 Sequence* candidateSeq = new Sequence(iss); m->gobble(iss);
1251 string candidateAligned = candidateSeq->getAligned();
1253 if (candidateSeq->getName() != "") { //incase there is a commented sequence at the end of a file
1255 if (candidateSeq->getAligned().length() != templateSeqsLength) {
1256 m->mothurOut(candidateSeq->getName() + " is not the same length as the template sequences. Skipping."); m->mothurOutEndLine();
1260 chimera->getChimeras(candidateSeq);
1262 if (m->control_pressed) { delete candidateSeq; return 1; }
1264 //if you are not chimeric, then check each half
1265 data_results wholeResults = chimera->getResults();
1267 //determine if we need to split
1268 bool isChimeric = false;
1270 if (wholeResults.flag == "yes") {
1271 string chimeraFlag = "no";
1272 if( (wholeResults.results[0].bsa >= minBS && wholeResults.results[0].divr_qla_qrb >= divR)
1274 (wholeResults.results[0].bsb >= minBS && wholeResults.results[0].divr_qlb_qra >= divR) ) { chimeraFlag = "yes"; }
1277 if (chimeraFlag == "yes") {
1278 if ((wholeResults.results[0].bsa >= minBS) || (wholeResults.results[0].bsb >= minBS)) { isChimeric = true; }
1282 if ((!isChimeric) && trimera) {
1283 //split sequence in half by bases
1284 string leftQuery, rightQuery;
1285 Sequence tempSeq(candidateSeq->getName(), candidateAligned);
1286 divideInHalf(tempSeq, leftQuery, rightQuery);
1288 //run chimeraSlayer on each piece
1289 Sequence* left = new Sequence(candidateSeq->getName(), leftQuery);
1290 Sequence* right = new Sequence(candidateSeq->getName(), rightQuery);
1293 chimera->getChimeras(left);
1294 data_results leftResults = chimera->getResults();
1296 chimera->getChimeras(right);
1297 data_results rightResults = chimera->getResults();
1299 //if either piece is chimeric then report
1300 Sequence trimmed = chimera->print(outMPI, outAccMPI, leftResults, rightResults);
1302 string outputString = ">" + trimmed.getName() + "\n" + trimmed.getAligned() + "\n";
1304 //write to accnos file
1305 int length = outputString.length();
1306 char* buf2 = new char[length];
1307 memcpy(buf2, outputString.c_str(), length);
1309 MPI_File_write_shared(outFastaMPI, buf2, length, MPI_CHAR, &status);
1313 delete left; delete right;
1317 Sequence trimmed = chimera->print(outMPI, outAccMPI);
1320 string outputString = ">" + trimmed.getName() + "\n" + trimmed.getAligned() + "\n";
1322 //write to accnos file
1323 int length = outputString.length();
1324 char* buf2 = new char[length];
1325 memcpy(buf2, outputString.c_str(), length);
1327 MPI_File_write_shared(outFastaMPI, buf2, length, MPI_CHAR, &status);
1334 delete candidateSeq;
1337 if((i+1) % 100 == 0){ cout << "Processing sequence: " << (i+1) << endl; m->mothurOutJustToLog("Processing sequence: " + toString(i+1) + "\n"); }
1340 if(num % 100 != 0){ cout << "Processing sequence: " << num << endl; m->mothurOutJustToLog("Processing sequence: " + toString(num) + "\n"); }
1345 catch(exception& e) {
1346 m->errorOut(e, "ChimeraSlayerCommand", "driverMPI");
1352 /**************************************************************************************************/
1354 int ChimeraSlayerCommand::createProcesses(string outputFileName, string filename, string accnos, string fasta) {
1358 int numNoParents = 0;
1361 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
1362 //loop through and create all the processes you want
1363 while (process != processors) {
1367 processIDS.push_back(pid); //create map from line number to pid so you can append files in correct order later
1369 }else if (pid == 0){
1370 num = driver(lines[process], outputFileName + toString(getpid()) + ".temp", filename, accnos + toString(getpid()) + ".temp", fasta + toString(getpid()) + ".temp");
1372 //pass numSeqs to parent
1374 string tempFile = outputFileName + toString(getpid()) + ".num.temp";
1375 m->openOutputFile(tempFile, out);
1376 out << num << '\t' << chimera->getNumNoParents() << endl;
1380 m->mothurOut("[ERROR]: unable to spawn the necessary processes."); m->mothurOutEndLine();
1381 for (int i = 0; i < processIDS.size(); i++) { kill (processIDS[i], SIGINT); }
1386 //force parent to wait until all the processes are done
1387 for (int i=0;i<processors;i++) {
1388 int temp = processIDS[i];
1392 for (int i = 0; i < processIDS.size(); i++) {
1394 string tempFile = outputFileName + toString(processIDS[i]) + ".num.temp";
1395 m->openInputFile(tempFile, in);
1396 if (!in.eof()) { int tempNum = 0; int tempNumParents = 0; in >> tempNum >> tempNumParents; num += tempNum; numNoParents += tempNumParents; }
1397 in.close(); m->mothurRemove(tempFile);
1401 //////////////////////////////////////////////////////////////////////////////////////////////////////
1402 //Windows version shared memory, so be careful when passing variables through the slayerData struct.
1403 //Above fork() will clone, so memory is separate, but that's not the case with windows,
1404 //////////////////////////////////////////////////////////////////////////////////////////////////////
1406 vector<slayerData*> pDataArray;
1407 DWORD dwThreadIdArray[processors];
1408 HANDLE hThreadArray[processors];
1410 //Create processor worker threads.
1411 for( int i=0; i<processors; i++ ){
1412 string extension = toString(i) + ".temp";
1413 slayerData* tempslayer = new slayerData((outputFileName + extension), (fasta + extension), (accnos + extension), filename, templatefile, search, blastlocation, trimera, trim, realign, m, lines[i]->start, lines[i]->end, ksize, match, mismatch, window, minSimilarity, minCoverage, minBS, minSNP, parents, iters, increment, numwanted, divR, priority, i);
1414 pDataArray.push_back(tempslayer);
1415 processIDS.push_back(i);
1417 //MySlayerThreadFunction is in header. It must be global or static to work with the threads.
1418 //default security attributes, thread function name, argument to thread function, use default creation flags, returns the thread identifier
1419 hThreadArray[i] = CreateThread(NULL, 0, MySlayerThreadFunction, pDataArray[i], 0, &dwThreadIdArray[i]);
1422 //Wait until all threads have terminated.
1423 WaitForMultipleObjects(processors, hThreadArray, TRUE, INFINITE);
1425 //Close all thread handles and free memory allocations.
1426 for(int i=0; i < pDataArray.size(); i++){
1427 num += pDataArray[i]->count;
1428 numNoParents += pDataArray[i]->numNoParents;
1429 CloseHandle(hThreadArray[i]);
1430 delete pDataArray[i];
1433 if (num == numNoParents) { m->mothurOut("[WARNING]: megablast returned 0 potential parents for all your sequences. This could be due to formatdb.exe not being setup properly, please check formatdb.log for errors."); m->mothurOutEndLine(); }
1435 rename((outputFileName + toString(processIDS[0]) + ".temp").c_str(), outputFileName.c_str());
1436 rename((accnos + toString(processIDS[0]) + ".temp").c_str(), accnos.c_str());
1437 if (trim) { rename((fasta + toString(processIDS[0]) + ".temp").c_str(), fasta.c_str()); }
1439 //append output files
1440 for(int i=1;i<processIDS.size();i++){
1441 m->appendFiles((outputFileName + toString(processIDS[i]) + ".temp"), outputFileName);
1442 m->mothurRemove((outputFileName + toString(processIDS[i]) + ".temp"));
1444 m->appendFiles((accnos + toString(processIDS[i]) + ".temp"), accnos);
1445 m->mothurRemove((accnos + toString(processIDS[i]) + ".temp"));
1448 m->appendFiles((fasta + toString(processIDS[i]) + ".temp"), fasta);
1449 m->mothurRemove((fasta + toString(processIDS[i]) + ".temp"));
1456 catch(exception& e) {
1457 m->errorOut(e, "ChimeraSlayerCommand", "createProcesses");
1462 /**************************************************************************************************/
1464 int ChimeraSlayerCommand::divideInHalf(Sequence querySeq, string& leftQuery, string& rightQuery) {
1467 string queryUnAligned = querySeq.getUnaligned();
1468 int numBases = int(queryUnAligned.length() * 0.5);
1470 string queryAligned = querySeq.getAligned();
1471 leftQuery = querySeq.getAligned();
1472 rightQuery = querySeq.getAligned();
1476 for (int i = 0; i < queryAligned.length(); i++) {
1478 if (isalpha(queryAligned[i])) {
1483 if (baseCount >= numBases) { leftSpot = i; break; } //first half
1486 //blank out right side
1487 for (int i = leftSpot; i < leftQuery.length(); i++) { leftQuery[i] = '.'; }
1489 //blank out left side
1490 for (int i = 0; i < leftSpot; i++) { rightQuery[i] = '.'; }
1495 catch(exception& e) {
1496 m->errorOut(e, "ChimeraSlayerCommand", "divideInHalf");
1500 /**************************************************************************************************/
1501 map<string, int> ChimeraSlayerCommand::sortFastaFile(string fastaFile, string nameFile) {
1503 map<string, int> nameAbund;
1505 //read through fastafile and store info
1506 map<string, string> seqs;
1508 m->openInputFile(fastaFile, in);
1512 if (m->control_pressed) { in.close(); return nameAbund; }
1514 Sequence seq(in); m->gobble(in);
1515 seqs[seq.getName()] = seq.getAligned();
1521 vector<seqPriorityNode> nameMapCount;
1522 int error = m->readNames(nameFile, nameMapCount, seqs);
1524 if (m->control_pressed) { return nameAbund; }
1526 if (error == 1) { m->control_pressed = true; return nameAbund; }
1527 if (seqs.size() != nameMapCount.size()) { m->mothurOut( "The number of sequences in your fastafile does not match the number of sequences in your namefile, aborting."); m->mothurOutEndLine(); m->control_pressed = true; return nameAbund; }
1529 sort(nameMapCount.begin(), nameMapCount.end(), compareSeqPriorityNodes);
1531 string newFasta = fastaFile + ".temp";
1533 m->openOutputFile(newFasta, out);
1535 //print new file in order of
1536 for (int i = 0; i < nameMapCount.size(); i++) {
1537 out << ">" << nameMapCount[i].name << endl << nameMapCount[i].seq << endl;
1538 nameAbund[nameMapCount[i].name] = nameMapCount[i].numIdentical;
1542 rename(newFasta.c_str(), fastaFile.c_str());
1547 catch(exception& e) {
1548 m->errorOut(e, "ChimeraSlayerCommand", "sortFastaFile");
1552 /**************************************************************************************************/
1553 map<string, int> ChimeraSlayerCommand::sortFastaFile(vector<Sequence>& thisseqs, map<string, string>& nameMap, string newFile) {
1555 map<string, int> nameAbund;
1556 vector<seqPriorityNode> nameVector;
1558 //read through fastafile and store info
1559 map<string, string> seqs;
1561 for (int i = 0; i < thisseqs.size(); i++) {
1563 if (m->control_pressed) { return nameAbund; }
1565 map<string, string>::iterator itNameMap = nameMap.find(thisseqs[i].getName());
1567 if (itNameMap == nameMap.end()){
1568 m->control_pressed = true;
1569 m->mothurOut("[ERROR]: " + thisseqs[i].getName() + " is in your fastafile, but is not in your namesfile, please correct."); m->mothurOutEndLine();
1571 int num = m->getNumNames(itNameMap->second);
1573 seqPriorityNode temp(num, thisseqs[i].getAligned(), thisseqs[i].getName());
1574 nameVector.push_back(temp);
1578 //sort by num represented
1579 sort(nameVector.begin(), nameVector.end(), compareSeqPriorityNodes);
1581 if (m->control_pressed) { return nameAbund; }
1583 if (thisseqs.size() != nameVector.size()) { m->mothurOut( "The number of sequences in your fastafile does not match the number of sequences in your namefile, aborting."); m->mothurOutEndLine(); m->control_pressed = true; return nameAbund; }
1586 m->openOutputFile(newFile, out);
1588 //print new file in order of
1589 for (int i = 0; i < nameVector.size(); i++) {
1590 out << ">" << nameVector[i].name << endl << nameVector[i].seq << endl;
1591 nameAbund[nameVector[i].name] = nameVector[i].numIdentical;
1598 catch(exception& e) {
1599 m->errorOut(e, "ChimeraSlayerCommand", "sortFastaFile");
1603 /**************************************************************************************************/